A discharge lamp used to backlight an LCD panel such as a cold cathode fluorescent lamp (CCFL) has terminal voltage characteristics that vary depending upon the immediate history and the frequency of a stimulus (AC signal) applied to the lamp. Until the CCFL is "struck" or ignited, the lamp will not conduct a current with an applied terminal voltage that is less than the strike voltage, e.g., the terminal voltage must be equal to or greater than 1500 Volts. Once an electrical arc is struck inside the CCFL, the terminal voltage may fall to a run voltage that is approximately 1/3 the value of the strike voltage over a relatively wide range of input currents. For example, the run voltage could be 500 Volts over a range of 500 microAmps to 6 milliAmnps for a CCFL that has a strike voltage of 1,500 Volt. When the CCFL is driven by an AC signal at a relatively low frequency, the CCFL's electrical arc tends to extinguish and ignite on every cycle, which causes the lamp to exhibit a negative resistance terminal characteristic. However, when the CCFL is driven by another AC signal at a relatively high frequency, the CCFL (once struck) will not extinguish on each cycle and will exhibit a positive resistance terminal characteristic. Since the CCFL efficiency improves at the relatively higher frequencies, the CCFL is usually driven by AC signals having frequencies that range from 50 KiloHertz to 100 KiloHertz.
Also, the mean time between failure for a CCFL is dependent upon several aspects of the operating environment. For example, driving the CCFL at a power level that is higher than the rated power level tends to shorten the useful lifetime of the lamp. Also, driving the CCFL with an AC signal that has a high crest factor can cause premature failure of the lamp. The crest factor is the ratio of the peak current to the average current that flows through the CCFL. Additionally, it is known that driving a CCFL with a relatively high frequency square-shaped AC signal will produce the maximum useful lifetime for the lamp. However, since the square shape of an AC signal may cause significant interference with another circuit disposed in the immediate vicinity of the circuitry driving the CCFL, the lamp is typically driven with an AC signal that has a less than optimal shape such as a sine-shaped AC signal.
Most small CCFLs are used in battery powered systems, e.g., notebook computers and personal digital assistants. The system battery supplies a direct current (DC) voltage ranging from 7 to 20 Volts with a nominal value of about 12V to an input of a DC to AC inverter. A common technique for converting a relatively low DC input voltage to a higher AC output voltage is to chop up the DC input signal with power switches, filter out the harmonic signals produced by the chopping, and output a relatively clean sine-shaped AC signal. The voltage of the AC signal is stepped up with a tansformer to a relatively high voltage, e.g., from 12 to 1500 Volts. The power switches may be bipolar junction transistors (BJT) or Field Effect Transistors (FET or MOSFET). Also, the transistors may be discrete or integrated into the same package as the control circuitry for the DC to AC converter.
Since resistive components tend to dissipate power and reduce the overall efficiency of a circuit, a typical harmonic filter for a DC to AC converter employs inductive and capacitive components that are selected to minimize power loss, i.e., each of the selected components should have a high Q value. The Q value identifies the "quality factor" of an inductor or a capacitor by indicating the ratio of energy stored to energy lost in the component for a complete cycle of an AC signal at a rated operational frequency. The Q value of a component will vary with the frequency and amplitude of a signal, so a filter must be designed for minimum (or acceptable) loss at the operating frequency and required power level. Also, some DC to AC converter filters incorporate the inductance of the step-up transformer, either in the magnetizing inductance of the primary or in the leakage inductance of the secondary.
A second-order resonant filter formed with inductive and capacitive components is also referred to as a "tank" circuit because the tank stores energy at a particular frequency. The unloaded Q value of the tank may be determined by measuring the parasitic losses of the tank components, i.e., the total energy stored by the tank for each cycle of the AC signal is divided by the total energy lost in the tank components each cycle. A high efficiency tank circuit will have a high unloaded Q value, i.e., the tank will employ relatively low loss capacitors and inductors.
The loaded Q value of a tank circuit may be measured when power is transferred through the tank from an energy source to. a load, i.e., the ratio of the total energy stored by the tank in each cycle of the AC signal divided by the total energy lost in the tank plus the energy transferred to the load in each cycle. The efficacy of the tank circuit as a filter depends on its loaded Q value, i.e., the higher the loaded Q value, the purer the shape of the sine wave output. Also, the efficiency of the tank circuit as a power transmitter depends on the ratio of the unloaded Q to the loaded Q. A high efficiency tank circuit will have an unloaded Q set as high as practical with a loaded Q set as low as possible. Additionally, the loaded Q of the tank circuit may be set even smaller to increase the efficiency of the filter, if the signal inputted to the tank has most of its energy in a fundamental frequency and only a small amount of energy is present in the lower harmonic frequencies.
The energy of a periodic waveform may be assigned to discrete frequencies, i.e., the fundamental repetition frequency and integer multiples of the fundamental repetition frequency. The fundamental repetition frequency is referred to as the fundamental and the integer multiples are termed harmonics. Generally, waveforms with sharp edges have fast rise and fall times and they have more energy in high order harmonics than waveforms with smooth edges and relatively slow transitions. However, generating waveforms with smooth, slow transitions usually causes fairly high power dissipation in the chopping switches, so the actual waveform is usually a compromise between efficient (sharp), fast edges and quiet (smooth), slow edges. Waveforms that are symmetric, i.e., the up-going waveform shape is the mirror image of the down-going shape but shifted in time, tends to suppress or cancel the even harmonics, which are the fundamental frequency times the integer values of 2, 4, 6, 8, 10, etc. The suppression or cancellation of the even harmonics is important because the harmonic frequency closest to the fundamental frequency is the second harmonic, which is the most difficult harmonic frequency to filter out of the waveform.
The largest component in a small DC to AC inverter circuit for a CCFL is the step-up transformer. Typically, this transformer includes a primary and a secondary winding coiled around a plastic bobbin mounted to a ferrite core. This type of transformer has two characteristic inductances associated with each winding, i.e., a magnetizing inductance and a leakage inductance. The value of the magnetizing inductance for each winding is measured when the other winding is configured as an open circuit, i.e., a no load state. Also, the value of the leakage inductance for each winding is measured when the other winding is configured as a short circuit.
The magnetizing inductance of a winding is a measure of how well the particular winding is coupled to the core of the transformer, i.e., a large magnetizing inductance is an indication that the magnetic flux of the winding is mostly contained within the core. A gap in the core will lower the magnetizing inductance because all of the magnetic flux is forced to leave the core at the gap. Thus, a relatively small transformer may be used to deliver a given power level, if the core is not gapped.
The leakage inductance is a measure of how poorly a winding is coupled to the other winding, i.e., a large leakage inductance indicates when the other winding is shorted. Since a high voltage, e.g., a strike voltage of 1500 volts, may be impressed on the secondary winding of the transformer for a CCFL converter, relatively thick insulators are typically used between the primary and the secondary windings. However, thick insulators tend to cause the leakage inductances of the primary and secondary windings to be relatively large.
The intensity of light emitted by a CCFL may be dimmed by driving the lamp with a lower power level (current). Dimming the light emitted by the CCFL enables the user to accommodate a wide range of ambient light conditions. Because the CCFL impedance will increase as the power level driving the lamp is reduced, i.e., an approximately constant voltage with a decreasing current, currents in the stray capacitances between neighboring conductors (e.g., ground shields, wiring) and the lamp tend to become significant. For example, if the control circuitry requires that one terminal of the CCFL is tied to signal ground for measuring current through the lamp, the current in the grounded terminal of the lamp will be significantly less than the current flowing into the other terminal of the lamp. In this case, a thermometer effect on the CCFL will be produced, whereby the grounded end of the lamp has almost no current flowing in it and the arc essentially extinguishes while the other end of the lamp is still arcing and emitting light. The thermometer effect may be greatly reduced by the technique of driving the CCFL, so that the signal at one end of the lamp is equal to and exactly out of phase with the signal at the other end. This technique is typically termed a balanced drive and it may be approximated by driving the CCFL with a floating secondary winding, i.e., neither end of the secondary winding is tied to ground.